|
|
 |
| 3-level MLM with version 6.11 |
 
|
| Message/Author |
|
|
| Brent Lyons posted on Wednesday, April 24, 2013 - 10:52 pm
|
|
|
I'm using v 6.11 with three levels of data
Lvl 3 - Person ("ID")
Lvl 2 - 2 relationships within person ("RelID"), manipulation is 0 or 1 ("RelCode")
Lvl 1 - within relationship, time series 10 time points (ex. "helpr")
I intend to do a path analysis assessing lvl2 effects on lvl1 associations, controlling for lvl3. How do I select clusters so that lvl2 (RelID) is nested within lvl-3 (ID). The data is organized in long format. Here is example input:
VARIABLE:
NAMES= RelID ID RelCode helpr helpe pa;
Missing are all (-9999);
USEVARIABLES =RelCode helpr pa harmr helpe;
CATEGORICAL ARE RelCode;
CLUSTER = RelID ID;
Within = helpr pa harmr;
Between = RelCode;
ANALYSIS: TYPE = twolevel complex random;
MODEL:
%within%
H | helpe ON helpr;
P | helpe ON pa;
M | helpe ON harmr;
%between%
H P M helpe;
H ON RelCode;
P ON RelCode;
M ON RelCode;
OUTPUT: sampstat;
Error message:
"One or more between-level variables have variation within a cluster for one or more clusters. Check your data and format statement.
Between Cluster ID with variation in this variable
Variable (only one cluster ID will be listed)"
My data is formatted properly - is a problem with my clusters?
Thanks, |
|
|
|
|
| If you can't see the problem with the data, send the input, data, output, and your license number to support@statmodel,com. |
|
|
| Back to top |
|
|
